A novel prediction method for lymph node involvement in endometrial cancer: machine learning

ObjectiveThe necessity of lymphadenectomy and the prediction of lymph node involvement (LNI) in endometrial cancer (EC) have been hotly-debated questions in recent years. Machine learning is a broad field that can produce results and estimations. In this study we constructed prediction models for EC patients using the Naïve Bayes machine learning algorithm for LNI prediction.MethodsThe study assessed 762 patients with EC. Algorithm models were based on the following histopathological factors: V1: final histology; V2: presence of lymphovascular space invasion (LVSI); V3: grade; V4: tumor diameter; V5: depth of myometrial invasion (MI); V6: cervical glandular stromal invasion (CGSI); V7: tubal or ovarian involvement; and V8: pelvic LNI. Logistic regression analysis was also used to evaluate the independent factors affecting LNI.ResultsThe mean age of patients was 59.1 years. LNI was detected in 102 (13.4%) patients. Para-aortic LNI (PaLNI) was detected in 54 (7.1%) patients, of which four patients had isolated PaLNI. The accuracy rate of the algorithm models was found to be between 84.2% and 88.9% and 85.0% and 97.6% for LNI and PaLNI, respectively. In multivariate analysis, the histologic type, LVSI, depth of MI, and CGSI were independently and significantly associated with LNI (p<0.001 for all).ConclusionsMachine learning may have a place in the decision tree for the management of EC. This is a preliminary report about the use of a new statistical technique. Larger studies with the addition of sentinel lymph node status, laboratory findings, or imaging results with machine learning algorithms may herald a new era in the management of EC.

Download Full-text

Comparison of MRI findings with actual HPE findings in case of carcinoma endometrium

10.1055/s-0039-1685340 ◽

2016 ◽

Author(s):

Shaveta Gupta

Keyword(s):

Endometrial Cancer ◽

Lymph Node ◽

Myometrial Invasion ◽

Lymph Node Involvement ◽

Common Finding ◽

Mri Findings ◽

Pelvic Mri ◽

Histopathological Findings ◽

Node Involvement ◽

Cervical Involvement

Objectives: The objectives of this study is to investigate the correlation of magnetic resonance imaging (MRI) in predicting the depth of myometrial invasion, cervical involvement and lymph node involvement and actual histopathological findings in the women with endometrial cancer. Methods: This is a reterospective study of the patients of endometrial cancer from Nov 2011 to Jan 2016 who underwent Surgery (Total abdominal Hystrectomy with B/l salpingoophorectomy with peritoneal washings with b/l pelvic lymphadenectomy with or without para aortic lymphadenectomy) at our centre Max Superspeciality Hospital. CE MRI Pelvis has been done pre operatively in every patient. After the surgery Histopathological reports of the specimen checked and compared with MRI findings of that case. The purpose of the study is to evaluate the validity of MRI findings of endometrial cancer in comparison to final histopathological findings. Results: For the detection of myometrial invasion, overall sensitivity of MRI is 93.9%, specificity is 66.6%, for cervical involvement Senstivity is 60% and specificity 1s 93.75% and for detection of lymph node involvement sensitivity is 66.6% and specificity is 93.5%. Most common Finding on MRI is thickened endometrium with disruption of Junction jone. Conclusions: Preoperative pelvic MRI is a sensitive method of identifying invasion to the myometrium in endometrial cancer. MRI Is a sensitive noninvasive modality in predicting locoregional spread in ca endometrium. Senstivity in detecting Myometrial invasion is high but sensitivity is less in detecting cervical involvement and lymph node involvement is less.

Download Full-text

Prognostic factors of lymph node involvement in endometrioid endometrial cancer: A SEER analysis

Gynecologic Oncology ◽

10.1016/j.ygyno.2014.03.347 ◽

2014 ◽

Vol 133 ◽

pp. 133

Author(s):

R. Vargas ◽

J.A. Rauh-Hain ◽

J.T. Clemmer ◽

R.M. Clark ◽

A. Goodman ◽

...

Keyword(s):

Endometrial Cancer ◽

Prognostic Factors ◽

Lymph Node ◽

Lymph Node Involvement ◽

Node Involvement ◽

Endometrioid Endometrial Cancer

Download Full-text

An axilla scoring system to predict non-sentinel lymph node status in breast cancer patients with sentinel lymph node involvement

Breast Cancer Research and Treatment ◽

10.1007/s10549-004-5781-z ◽

2005 ◽

Vol 91 (2) ◽

pp. 113-119 ◽

Cited By ~ 184

Author(s):

Emmanuel Barranger ◽

Charles Coutant ◽

Antoine Flahault ◽

Yann Delpech ◽

Emile Darai ◽

...

Keyword(s):

Breast Cancer ◽

Lymph Node ◽

Sentinel Lymph Node ◽

Cancer Patients ◽

Scoring System ◽

Lymph Node Involvement ◽

Lymph Node Status ◽

Breast Cancer Patients ◽

Node Involvement ◽

Sentinel Lymph Node Status

Download Full-text

Detection of Lymphovascular Invasion by D2-40 (Podoplanin) Immunoexpression in Endometrial Cancer

International Journal of Gynecological Cancer ◽

10.1097/igc.0b013e318269139b ◽

2012 ◽

Vol 22 (8) ◽

pp. 1442-1448 ◽

Cited By ~ 25

Author(s):

Sarah K. Weber ◽

Axel Sauerwald ◽

Martin Pölcher ◽

Michael Braun ◽

Manuel Debald ◽

...

Keyword(s):

Overall Survival ◽

Endometrial Cancer ◽

Lymph Node ◽

Lymph Node Metastasis ◽

Lymphovascular Invasion ◽

Lymph Node Involvement ◽

Lymph Vessel ◽

Node Metastasis ◽

Lymph Vessel Invasion ◽

Node Involvement

BackgroundLymph node involvement is a major feature in tumor spread of endometrial cancer and predicts prognosis. Therefore, evaluation of lymph vessel invasion (LVI) in tumor tissue as a predictor for lymph node metastasis is of great importance. Immunostaining of D2-40 (podoplanin), a specific marker for lymphatic endothelial cells, might be able to increase the detection rate of LVI compared with conventional hematoxylin-eosin (H-E) staining. The aim of this retrospective study was to analyze the eligibility of D2-40–based LVI evaluation for the prediction of lymph node metastases and patients’ outcome.Patients and MethodsImmunohistochemical staining with D2-40 monoclonal antibodies was performed on paraffin-embedded tissue sections of 182 patients with primary endometrioid adenocarcinoma treated in 1 gynecologic cancer center. Tumors were screened for the presence of LVI. Correlations with clinicopathological features and clinical outcome were assessed.ResultsImmunostaining of D2-40 significantly increased the frequency LVI detection compared with conventional H-E staining. Lymph vessel invasion was identified by D2-40 in 53 (29.1%) of 182 tumors compared with 34 (18.3%) of 182 carcinomas by routine H-E staining (P = 0.001). D2-40 LVI was detectable in 81.0% (17/21) of nodal-positive tumors and significantly predicted lymph node metastasis (P = 0.001). Furthermore, D2-40 LVI was an independent prognostic factor for patients overall survival considering tumor stage, lymph node involvement, and tumor differentiation (P < 0.01). D2-40–negative tumors confined to the inner half of the myometrium showed an excellent outcome (5-year overall survival, 97.8%).ConclusionsD2-40–based LVI assessment improves the histopathological detection of lymphovascular invasion in endometrial cancer. Furthermore, LVI is of prognostic value and predicts lymph node metastasis. D2-40 LVI detection might help to select endometrial cancer patients who will benefit from a lymphadenectomy.

Download Full-text

The value of tumor marker and cytokine analysis for the assessment of regional lymph node status in cervical cancer patients

International Journal of Gynecological Cancer ◽

10.1111/j.1525-1438.2007.01176.x ◽

2008 ◽

Vol 18 (6) ◽

pp. 1279-1284 ◽

Cited By ~ 11

Author(s):

B. Kotowicz ◽

M. Fuksiewicz ◽

M. Kowalska ◽

J. Jonska-Gmyrek ◽

M. Bidzinski ◽

...

Keyword(s):

Cervical Cancer ◽

Lymph Node ◽

Lymph Node Metastases ◽

Regional Lymph Node ◽

Serum Levels ◽

Lymph Node Involvement ◽

Lymph Node Status ◽

Node Status ◽

Node Involvement ◽

Node Metastases

The aim of the study was to evaluate the utility of the measurements of the circulating tumor markers, squamous cell carcinoma antigen (SCCA), CA125, carcinoembryonic antigen (CEA), cytokeratin fragment 19 (CYFRA 21.1), and the cytokines, interleukin-6 and vascular endothelial growth factor (VEGF), to estimate regional lymph node involvement in patients with cervical cancer. The study comprised 182 untreated patients with cervical cancer. The regional lymph node status was assessed either by the postsurgical histopathologic examination or by the computed tomography (CT). Concentrations of SCCA, CEA, and CA125 were determined using the Abbott Instruments system, of CYFRA 21.1 by the Roche kits, and of IL-6 and VEGF by the ELISA of R&D Systems (Minneapolis, MN). For the statistical analyses, Mann–Whitney U test and χ2 test were applied. Serum levels of SCCA, CEA, CA125, CYFRA 21.1, IL-6, and VEGF were measured in patients with specified pelvic and para-aortic lymph node status. SCCA, CA125, and IL-6 levels were found to be significantly higher in patients with lymph node metastases than in those with no lymph node involvement. Also, the percentage of patients with simultaneously elevated concentrations of SCCA and CA125 or SCCA and IL-6 differed depending on the lymph node status and was significantly higher in the series of patients with lymph node metastases. Simultaneous assessment of serum levels of SCCA and CA125 or SCCA and IL-6 in patients with cervical cancer may be useful for the regional lymph node evaluation, especially in patients with advanced stages, when the lymph nodes are examined only by CT, with no histologic confirmation.

Download Full-text

Comparison of three different risk-stratification models for predicting lymph node involvement in endometrioid endometrial cancer clinically confined to the uterus

Journal of Gynecologic Oncology ◽

10.3802/jgo.2017.28.e78 ◽

2017 ◽

Vol 28 (6) ◽

Cited By ~ 8

Author(s):

Vakkas Korkmaz ◽

Mehmet Mutlu Meydanli ◽

Ibrahim Yalçın ◽

Mustafa Erkan Sarı ◽

Hanifi Sahin ◽

...

Keyword(s):

Endometrial Cancer ◽

Lymph Node ◽

Risk Stratification ◽

Lymph Node Involvement ◽

Node Involvement ◽

Endometrioid Endometrial Cancer

Download Full-text

Comparison of ER, PR, Ki67 and HER-2/neu Reactivity Pattern with Patients’ Age, Histologic Grade, Tumor Size and Lymph Node Status in Invasive Ductal Breast Cancer

International Journal of Cancer Management ◽

10.5812/ijcm.98325 ◽

2020 ◽

Vol 13 (6) ◽

Author(s):

Mahsa Ahadi ◽

Motahareh Heibatollahi ◽

Sara Zahedifard

Keyword(s):

Breast Cancer ◽

Lymph Node ◽

Tumor Size ◽

Histological Grade ◽

Lymph Node Involvement ◽

Lymph Node Status ◽

Invasive Ductal Breast Cancer ◽

Ductal Breast Cancer ◽

Node Involvement ◽

Her 2

Background: Breast cancer is the most prevalent neoplasm diagnosed in Iranian women. Objectives: The current study was performed to measure the hormone receptor status and its possible connection with the patient’s age, tumor size, histological grade, and lymph node status and involvement in patients with invasive ductal breast cancer (IDBC) Methods: A total of 103 women with IDBC recently diagnosed at the Department of Pathology of Shohada-E-Tajrish Hospital were entered into this study. The mean age of the patients was 48.4 years, and 59.2% of cases were 50 years old or less. Results: Most lesions (78.6%) were more than 2 cm at their greatest dimension. Grade-II lesions were observed in a large number of patients and 59.8% of cases had lymph node involvement. Positive ER, PR, and HER-2/neu were detected in 59%, 57%, and 29% of patients, respectively. A significant correlation was found between patients’ age and histologic score, tumor dimension and both histologic score and nuclear grade, and, finally, between lymph node involvement and nuclear grade. Conclusions: According to previous studies, the evaluation of hormone receptor status in patients with breast cancer is strongly recommended. Here, by studying its possible connection with the patient’s age, tumor size, histological grade, and lymph node metastasis, we detected some biomarkers, which could be used as prognostic indices in these patients. These biomarkers could help us in the clinical management of patients with IDBC by providing the best therapeutic options.

Download Full-text

Development of a Radiomic-Based Model Predicting Lymph Node Involvement in Prostate Cancer Patients

Cancers ◽

10.3390/cancers13225672 ◽

2021 ◽

Vol 13 (22) ◽

pp. 5672

Author(s):

Vincent Bourbonne ◽

Vincent Jaouen ◽

Truong An Nguyen ◽

Valentin Tissot ◽

Laurent Doucet ◽

...

Keyword(s):

Neural Network ◽

Prostate Cancer ◽

Lymph Node ◽

Prediction Model ◽

Risk Prediction ◽

Prediction Models ◽

Lymph Node Involvement ◽

Visual Interpretation ◽

Node Involvement ◽

Testing Set

Significant advances in lymph node involvement (LNI) risk modeling in prostate cancer (PCa) have been achieved with the addition of visual interpretation of magnetic resonance imaging (MRI) data, but it is likely that quantitative analysis could further improve prediction models. In this study, we aimed to develop and internally validate a novel LNI risk prediction model based on radiomic features extracted from preoperative multimodal MRI. All patients who underwent a preoperative MRI and radical prostatectomy with extensive lymph node dissection were retrospectively included in a single institution. Patients were randomly divided into the training (60%) and testing (40%) sets. Radiomic features were extracted from the index tumor volumes, delineated on the apparent diffusion coefficient corrected map and the T2 sequences. A ComBat harmonization method was applied to account for inter-site heterogeneity. A prediction model was trained using a neural network approach (Multilayer Perceptron Network, SPSS v24.0©) combining clinical, radiomic and all features. It was then evaluated on the testing set and compared to the current available models using the Receiver Operative Characteristics and the C-Index. Two hundred and eighty patients were included, with a median age of 65.2 y (45.3–79.6), a mean PSA level of 9.5 ng/mL (1.04–63.0) and 79.6% of ISUP ≥ 2 tumors. LNI occurred in 51 patients (18.2%), with a median number of extracted nodes of 15 (10–19). In the testing set, with their respective cutoffs applied, the Partin, Roach, Yale, MSKCC, Briganti 2012 and 2017 models resulted in a C-Index of 0.71, 0.66, 0.55, 0.67, 0.65 and 0.73, respectively, while our proposed combined model resulted in a C-Index of 0.89 in the testing set. Radiomic features extracted from the preoperative MRI scans and combined with clinical features through a neural network seem to provide added predictive performance compared to state of the art models regarding LNI risk prediction in PCa.

Download Full-text